J. Microbiol. Biotechnol. 2020. 30(8): 1169–1179 https://doi.org/10.4014/jmb.2003.03018 Differential Impacts on Bacterial Composition and Abundance in Rhizosphere Compartments between Al-Tolerant and Al-Sensitive Soybean Genotypes in Acidic Soil Zhong-Ling Wen1†, Min-Kai Yang1†, Aliya Fazal1†, Yong-Hui Liao1, Lin-Run Cheng2, Xiao-Mei Hua3, Dong-Qing Hu1, Ji-Sen Shi 4, Rong-Wu Yang1, Gui-Hua Lu1,4, Jin-Liang Qi1,4*, Zhi Hong1*, Qiu-Ping Qian2*, and Yong-Hua Yang1,4* 1Institute for Plant Molecular Biology, State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing 210023, P.R. China 2Jinhua Academy of Agricultural Sciences, Jinhua 321017, P.R. China 3Research Center for Soil Pollution Prevention and Control, Nanjing Institute of Environmental Sciences, MEE, Nanjing 210042, P.R. China 4Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry University, Nanjing 210037, P.R. China In this study, two soybean genotypes, i.e., aluminum-tolerant Baxi 10 (BX10) and aluminum- sensitive Bendi 2 (BD2), were used as plant materials and acidic red soil was used as growth medium. The soil layers from the inside to the outside of the root are: rhizospheric soil after washing (WRH), rhizospheric soil after brushing (BRH) and rhizospheric soil at two sides (SRH), respectively. The rhizosphere bacterial communities were analyzed by high-throughput sequencing of V4 Received: March 11, 2020 hypervariable regions of 16S rRNA gene amplicons via Illumina MiSeq. The results of alpha diversity Accepted: May 15, 2020 analysis showed that the BRH and SRH of BX10 were significantly lower in community richness than that of BD2, while the WRH exhibited no significant difference between BX10 and BD2. Among the First published online: May 20, 2020 three sampling compartments of the same soybean genotype, WRH had the lowest community richness and diversity while showing the highest coverage. Beta diversity analysis results displayed *Corresponding authors no significant difference for any compartment between the two genotypes, or among the three Y.H.Y. different sampling compartments for any same soybean genotype. However, the relative abundance Phone/Fax: +86-25-89686305 E-mail: [email protected] of major bacterial taxa, specifically nitrogen-fixing and/or aluminum-tolerant bacteria, was Q.P.Q. significantly different in the compartments of the BRH and/or SRH at phylum and genus levels, Phone: +86-25-89686305 indicating genotype-dependent variations in rhizosphere bacterial communities. Strikingly, as E-mail: [email protected] compared with BRH and SRH, the WRH within the same genotype (BX10 or BD2) always had an Z.H. enrichment effect on rhizosphere bacteria associated with nitrogen fixation. E-mail: [email protected] J.L.Q. Keywords: Soybean genotypes, rhizosphere compartments, bacterial communities, nitrogen fixation, E-mail: [email protected] aluminum tolerance †These authors contributed equally to this work. Supplementary data for this Introduction paper are available on-line only // Acidity in soil affects about 30% of the world’s total land area and 50% of the world’s potential arable land [1]. In at http: jmb.or.kr. South China, acidic red soil is a typical soil type with low pH value, low phosphorus content and high aluminum pISSN 1017-7825 toxicity, which are directly related to restrictive crop growth and yield [2-4]. In order to adapt to the stressful eISSN 1738-8872 conditions, plants can release root exudates such as organic acids, which can chelate toxic Al (aluminum) and mobilize P (phosphorus) [5-7]. Soybean is an important protein and oil source in the world and can also be Copyright© 2020 by utilized as an excellent rotation and intercropping crop [8]. Previous studies have shown that under severe P stress, The Korean Society for Al-tolerant soybean Baxi 10 (BX10) has a greater citrate efflux rate and nearly 2 times higher yield compared to Al- Microbiology and sensitive soybean Bendi 2 (BD2) [9]. Moreover, the rhizosphere soil of BX10 exhibited changes in the ratio of Biotechnology gram-negative/gram-positive bacteria compared with BD2 [10]. Each plant species has its own specific rhizospheric bacterial community as different rhizosphere bacteria A 2020 ⎪ Vol. 30 ⎪ No. 8 1170 Wen et al. respond differently to different compounds of root exudates [11, 12]. Previous studies have reported that plants can strongly influence the composition, structure, and activity of rhizosphere microbiota, especially with respect to the active populations [13-18]. Moreover, rhizosphere microorganisms such as bacteria, are closely related to plant health and growth [13, 19-21], thus leading to a complex binary relationship between plants and their root- associated microorganisms. For example, experiments conducted by Li et al. showed that there are differences in the diversity and abundance of functional nifH gene (coding for nitrogenase iron protein) of the rhizosphere bacteria community at different growth stages between BX10 and BD2 [10, 22]. In this study, we aimed to elucidate the impacts of Al-tolerant (BX10) and Al-sensitive (BD2) soybean genotypes on the composition and abundance of rhizosphere bacterial communities at the flowering stage, among three different sampling compartments in acidic soil, via Illumina MiSeq sequencing platform. Materials and Methods Plant Materials, Sampling Methods and DNA Extraction In this study, two soybean genotypes(Glycine max (L.) Merr.), BX10 (Baxi 10, aluminum (Al)-tolerance type) and BD2 (Bendi 2, Al-sensitive type), were selected as plant materials. Soybeans were planted in a rhizobox (rhizosphere box, 200 mm in length, 150 mm in width and 200 mm in depth) which was constructed with PVC material and divided into five parts by using a frame covered with nylon film, in order to prevent the root and root exudates from entering different soil compartments, but not water and nutrition [23, 24]. Finely sieved acidic red soil (pH 4.43) collected from the Ecological Experiment Station of Red Soil (28.208 N, 116.937 E) of the Chinese Academy of Sciences, Yingtan, Jiangxi Province, China [22], was evenly packed into five parts of the rhizobox while soybean was allowed to grow in the middle part. Sampling at flowering stage was performed as described previously by Li et al. with some modifications [25, 26]. The soil in the compartments on both sides of the rhizobox (rhizosphere soils between two partitions in rhizobox) was named SRH (rhizospheric soil at two sides) [24]. BRH (rhizospheric soil after brushing) samples were collected by brushing off the soil tightly adhering to the root surface, and then the WRH (rhizospheric soil after washing) samples were collected by centrifugation at 4,000 ×g for 10 min after being washed with phosphate- buffered saline (PBS). Finally, all the samples were stored at -80oC prior to DNA extraction. Metagenomic DNA of every biological replicate was extracted from approximately 0.30 g of soil by using the PowerSoil DNA Isolation Kit (MoBio Laboratories Inc., USA). The experimental method followed the instructions with minor modifications [27, 28]. After extraction, the quality of DNA samples was assessed on 1% agarose gel and quantified by using a Qubit Fluorometer (Qubit 2.0, Invitrogen, USA) to minimize the variability in surveys of bacterial communities [29]. 16S rDNA Amplicon Sequencing and Analysis We used an improved dual-index high-throughput sequencing with paired-end 250nt and amplicons of approximately 290 bp encompassing the V4 hypervariable region of the 16S rDNA [30] using the following primers: forward primer 515F (5’-GTGCCAGCMGCCGCGGTAA-3’) and reverse primer 806R (5’-GGACTA CHVGGGTWTCTAAT-3’) [31]. The concentration of each qualified metagenomic DNA was tested and ensured to be more than 0.4 ng/μl [29]. PCR amplification, product purification, library quality determination and high- throughput sequencing of the qualified libraries on the Illumina MiSeq platform (Illumina, USA) with MiSeq Reagent Kit were conducted by BGI Tech Solutions Co., Ltd. (China). A total of 18 sequencing clean data has been submitted to the Sequence Read Archive (SRA) and the SRA accession number is PRJNA574806. In order to characterize the bacterial community composition and structure in the SRH, BRH and WRH, high- throughput sequencing of 16S rDNA amplicons using V4 region on the Illumina MiSeq platform was performed. Then, the clean tags were clustered into Operational Taxonomic Units (OTUs) with a 97% similarity by using UPARSE software (v7.0.1090) [32] and the chimeras were filtered out by using UCHIME(v4.2.40) and 16S rDNA was screened for chimeras by mapping to a gold database (v20110519) [33]. The sub-sampling of OTU was performed using the software R (v3.1.3) in I-Sanger (http://www.i-sanger.com) according to the minimum sample number sequence. A total of 1,736,758 qualified paired-end reads with an average count of 96,486 (range: 80,603–114,407) per sample were obtained from 18 samples (Table S1). The high quality paired-end reads were then connected to tags based on 250 bp overlaps and a total of 1,214,537 tags were obtained (Table S1). After the removal of chimeras, clean tags were clustered into OTUs and a total of 35,651 OTUs were obtained (Table S1). The detailed information on OTUs with/without sub-sampling is summarized in Table S2 (Table S1). Then, the alpha and beta diversity analyses were conducted based on OTUs and species annotation results. A rank-abundance curve was used to explain species abundance and evenness, and a Venn diagram was used to count the number of common and unique OTUs in multiple samples with 97% similarity level. Pan species were the sum of all species contained in all samples while core species were the number of common species in all the samples. Alpha Diversity, Beta Diversity, Functional Prediction and Analysis, and Statistical Analysis In order to analyze the complexity of species diversity in the environment, we used alpha diversity to reflect the community richness (Sobs, Chao, and Ace indices), community diversity (Shannon and Simpson indices) and community coverage (coverage index), while beta diversity analysis was calculated by using QIIME (v 1.8.0).